The overall objective of this project is to identify new opportunities for the treatment of refractory prostate cancer using pathway-directed high-throughput screening of an unbiased combinatorial small molecule library. We will synthesize a library comprised of diverse molecular scaffolds and screen it using yeast-based assays in which growth is dependent upon activation of an androgen receptor mutant. These mutants will be derived from prostate cancers that have become refractory to antiandrogen therapies, including those with alterations in the ligand binding domain, transcriptional activation domain, hinge region, and DNA-binding domain. Small molecule inhibitors of signaling by these androgen receptor mutants may act by novel mechanisms, such as by interfering with binding to coactivators or chaperones, by preventing nuclear localization, dimerization, or DNA binding, or by stimulating association with repressors. Thus, these molecules may serve as new probes of androgen receptor function and, potentially, as lead compounds for the treatment of androgen receptor-dependent advanced prostate cancer. More generally, this is a potentially powerful approach to identifying new molecular targets for therapeutic intervention since these screens are directed toward an entire signaling pathway rather than a single biological target. This provides the opportunity to identify new proteins that are involved in the signaling pathway and also previously unrecognized binding sites on proteins already known to act in the pathway. However, because multiple targets are present, most of which have no ligand or structural basis for guiding rational design of a combinatorial library, an unbiased library approach is required. Since most libraries are currently synthesized around an individual core scaffold, their structural diversity is actually quite limited. Thus, the development of new strategies to provide diverse, multi-scaffold libraries is required. We will pursue these broad goals thorough a multidisciplinary, collaborative research project comprised of three specific aims: (1) Synthesize a small, multi-scaffold combinatorial library of 500-1000 compounds on an accelerated timeline of 6-12 months, (2) Develop yeast assays in which activation of mutant androgen receptors can be efficiently assessed, and (3) Carry out high throughput screening to identify novel inhibitors of AR function and activation for subsequent evaluation as biological probes and therapeutic lead compounds.